DNA HYBRIDIZATION IN THE GENUS DROSOPHILA1

THRESSA D. ENTINGH

Department of Biology, Uniuersity of Chicago2

Received December 15, 1969

EVOLUTIONARY trees have classically been based upon species character- istics which are phenotypic in nature. It is likely that direct measurement of

the number o r proportion of genes shared by different species will be of basic importance in understanding evolutionary relationships. A quite direct compari- son of a large number of genes of different organisms can be made through the nucleic acid hybridization technique (HOYER and ROBERTS 1968). When species relationships are discussed, phenetic relationships must be distinguished from phylogenetic relationships. Phenetic relationships are based on indices of overall similarity between species. Phylogenetic relationships purport to trace the gene- alogies of species, and are based on specific characteristics selected to produce a genealogy which is most compatible with all the phenotypic characteristics of the species. The purpose of the present experiments is to determine if the species relationships which are derived from DNA hybridization data are phenetic or phylogenetic in nature.

A large amount of information about DNA similarity relatianships among bacteria (MCCARTHY and BOLTON 1963) has been obtained, but in the absence of detailed morphological and paleontological data these studies have contributed little to our understanding of species relationships. On the other hand, only limited information is available from vertebrate DNA comparisons (HOYER et al. 1965), but that which exists is in general agreement with phylogenies derived from anatomical, embryological, and paleontological data. Since the phylogenetic and phenetic relationships of species in the genus Drosophila have been com- paratively well established (STURTEVANT and NOVITSKI 1941 ; PATTERSON and STONE 1952; THROCKMORTON 1962a, 1962b; HUBBY and THROCKMORTON 19651, DNA hybridization between a number of Drosophila species was studied in order to compare DNA similarity data with cytogenetic, biochemical, and morphologi- cal data. LAIRD and MCCARTHY (1968) have compared the DNA of three Dro- sophila species using the hybridization technique and found relatively large differences among these species, shawing that DNA hybridization can be used to distinguish between species in the same genus.

Not only have hybridization studies given insights into relationships between the DNAs of different organisms, but they have revealed some surprising proper- ties of DNA itself. For example, intraspecific DNA reassociation studies have shown that eukaryotic DNA contains a fraction which, following denaturation,

This research was supported in part by USPHS GM 11216 and GM-150 2 Present address Department of Biochemistry, University of North Carolina, Chapel Hill, North Carolina

Genetics 66: 55-68 September 1970

56 T. D. ENTINGH

TABLE 1

List of species and strains used

Drosophila species ~

Subgenus Collection No. Collection locality

1. victoria 2. lebanonensis 3. melanogaster 4. simulans 5. willistoni 6. paulistorum 7. virilis 8. virilis 9. borealis

IO. micromelanica 11. melanica 12. paramelanica’ 13. euronotus’ 14. nigromelanica 15. pengi 16. hydeii 17. fuluamacula 18. funebris 19. pallidipennis

Pholadoris Pholadoris. Sophophora Sop hop hora Sophophora Sophophora Drosophila Drosophila Drosophila Drosophila Drosophila Drosophila Drosophila Drosophila Drosophila Drosophila Drosophila Drosophila Drosophila

1865.3 1733.1 Oregon-R H231.2 2267.9 1975.21 1801.1 1736.7 2077.4L 2160.12 1720.3 28-12 45b 2551.3

2360.30 H435.7 2082.1 H191.48

. . . .

Nevo, Utah Beirut, Lebanon

Raratonga, British Guinea

Belem, Brazil Texmelucan, Mexico Hang Chow, China Itasca Park, Minnesota Madera Canyon, Arizona Cliff, New Mexico St. Louis, Missouri St. Louis, Missouri Cold Spring Harbor, N.Y. Japan Cave Creek, Arizona Latitia, Columbia Minneapolis, Minnesota Bucaramanga, Columbia

* Obtained from H. STALKER. Species classification is according to PATTERSON and STONE (1952) and THROCKMORTON (1962).

reanneals rapidly compared with the rest of the DNA (WALKER and MCCLAREN 1965; BRITTEN and KOHNE 1968). This DNA fraction might be localized in certain chromosome structures. Since D. melanoguster stocks with varying amounts of heterochromatin can be obtained, experiments were done to deter- mine if the fast annealing DNA is localized only in the heterochromatic portions of the genome.

MATERIALS AND METHODS

Stocks: The 19 species and strains used are listed in Table 1 along with the collection numbers and localities. Flies were raised at 23°C on either standard cornmeal-agar medium, CARPENTER’S (1950) medium, or banana-agar medium. Young adults were collected and stored at -23°C. D. melanogaster stocks of strains carrying varying amounts of X-centromeric heterochromatin were obtained from W. K. BAKER’S laboratory at the University of Chicago. In (I) sc4t~sc8R,yw/Y/ywf males lack the nucleolus organizer (NO) heterochromatin on the X but carry it on the Y chro- mosome, sc8 B Znwa scJ;sc1gL/cy cnz females carry two NO heterochromatic regions on each X chromosome. X/O males collected from the cross of Ys wyY1 y55flo/O males and Oregon-R virgin females carry one dose of NO heterochromatin on the X chromosome but have no Y chro- mosome.

Isotopic labeling: To label Drosophila DNA, 3H-methyl-thymidine (New England Nuclear, 17 C/mM) was added to either cornmeal or CARPENTER’S medium to a final concentration of 50 pc/ml. 3H-thymidine is a suitable label for yeast containing media since there is no incorporation of the label into the yeast DNA (LAIRD and MCCARTHY 1968). Adult females were allowed to lay eggs on this medium for 24 hr. Ten to eighteen days later emerging adults were collected and stored for extraction of DNA. Purified DNA was assayed for specific activity on nitrocellulose

D N A HYBRIDIZATION IN DROSOPHILA 57

filters (Schleicher and Schuell B-6) in a Nuclear-Chicago scintillation counter. Various prepa- rations yielded DNAs containing from 680 to 2,675 cpm/pg.

Zsolation of DNA: Drosophila DNA was extracted according to the procedures of RITOSSA and SPIEGELMAN (1965) and LAIRD and MCCARTHY (1968). Nucleic acid, protein, and poly- saccharide concentrations were estimated from optical density measurements scanned between 230 and 360 m p . The DNA concentration was determined by the DISCHE (1944) reaction and preparations containing more than 20% RNA were redigested with RNAase. The yield of purified DNA varied from 0.5-1.0 mg/g flies. E. coli DNA was extracted using the procedure of MARMUR (1961). All DNA solutions were stored a t 4°C over chloroform.

Hybridization: To prepare filterbound DNA, about 0.5 mg of purified DNA (0.10 mg/ml) was denatured in 0.1 >< SSC (1 >< SSC is 0 . 1 5 ~ sodium chloride, 0 . 0 1 5 ~ trisodium citrate) by heating to 95°C for 10 min. The denatured DNA was plunged into ice and diluted to 0.05 mg/ml with 3>< SSC and deposited on a membrane filter (Schleicher and Schuell B-6) which had been soaked in 3>< SSC for 3 min. The filter was washed with 100 ml of 3>< SSC, dried at room tem- perature overnight, and preincubated for 6 hr at 65°C in the medium specified by DENHARDT (1966). The filter was then dried in a vacuum oven at 80°C for two hours. The amount of DNA per filter was determined from OD,,, measurements before and after filtration. Labeled and unlabeled DNAs used for duplex reactions were denatured as above and sheared by passing them through a Tomac 1 ml syringe (GILLESPIE and SPIEGELMAN 1965). These DNAs were incubated with the filterbound DNA in scintillation vials at 65°C for 16 h r at a concentration of 4-7 pg in 0.15 ml of 3>< SSC. The filters were then washed with 41) ml of 3>< SSC, dried overnight, and counted in a liquid scintillation counter. The fraction of DNA which reanneals under these con- ditions is called the fast annealing fraction.

The remaining DNA fraction was dialyzed, concentrated by lyophilization, dissolved in dis- tilled water at 2-5 mg/ml and stored frozen. To initiate further hybridization this DNA fraction was adjusted to 3>< SSC, redenatured, sheared, placed in 1 m l 3 x SSC with 5 to 7 filters contain- ing immobilized DNA, and incubated with mild shaking at 65°C in silicone-treated vials for a specified number of days. This DNA fraction is called the slow annealing fraction.

RESULTS

1. Factors aflecting hybridization: The proportion of fragmented DNA which annealed to homologous filterbound

DNA reached a maximum of about 20% (Figure 1 ) at approximately 12 hr and remained constant for at least 72 hr. The lack of complete annealing did not appear to be due to the saturation of filterbound DNA since removal of the saturated filter and the insertion of new filterbound DNA led to only 3.8% additional annealing of the DNA fragments. Nor was the cessation of annealing due to significant reassociation of the DNA fragments themselves, since reheating the incubation mixture reinitiated only 4.8% further annealing. DNA which annealed under the above conditions is the fast annealing DNA. By varying the experimental conditions reassociation of a portion of the remaining DNA (slow fraction) can be obtained (Figure 2). 2. Nucleotide sequence similarity among diflerent Drosophila species:

a. Similarity of the fast annealing DNA fraction. All the following experi- ments were done with total (nonfractionated) DNA under the conditions which allow annealing of the fast DNA fraction. A competition method was used to determine the percentage of DNA sequences held in common by different pairs of species: radioactive DNA fragments were reacted with homologous filterbound DNA in the presence of increasing and excessive amounts of nonradioactive frag-

58 T. D. ENTINGH

1 J 0

20 - L

g c

D

z 10 2 ' 5 P 15: / 1 J 0 e

20 - I J I . JIJ L

b e

g e 15- c 0

D m

z 10- 2 ' 5 P

l imo of incubation (hrd

FIGURE 1.-The effect of different times of incubation on the proportion of DNA fragments bound. Fifty micrograms filterbound melanogaster DNA were placed in 1.15 ml 3>< SSC containing 5 pg 3H-melanogaster DNA (1576 cpm/pg) and were incubated at 65°C for specified time periods.

TABLE 2

The relatedness of several Drosophila species as measured by DNA competition interactions*

3H-DNA fragments D . D. D . D . virilis

Competitor DNA melanogaster willistoni lebanonensis D. funebris (1801.1) D . hydeii

D. melanogastert D. simulans D. willistoni D. paulistorum D. victoria D. lebanonensis D. funebris D. uirilis (1801.1) D. uirilis (1 736.7) D. borealis D. hydeii D. fuluamacula D. micromelanica D. pallidipennis

100 69.9 . . . 81.5 . . . . . . . . 100 60.6 . . 88.7 . .

69.3 65.0 92.7 . . . . . . 100

34.2 38.2 47.3 40.7 52.8 . . . 42.6 . . . . . . 39.0 . . . . . . . . . 35.0 46.6

28.0 29.8 53.9 29.7 . . . . . . . . . . . . . .

. . 38.1

38.7 51.9 55.2 59.9 . . . 55.6

100 52.5 . . . 100 . . . 100 . . . 73.3

. . . . . . 59.2 . . .

30.0 30.3 35.8 23.5

. . .

. . . 35.1 53.2

56.1

75.5

. . .

100

. . .

. . .

* Conditions of incubation of these competition experiments are as in legend to Figure 3. The specific activity of the DNAs varied from 9762,675 cpm/pg DNA. The mean standard error was 2.4%; the range of the standard errors 1.2-3.8%. Means were computed from three or four samples. t melanogaster-simulans, willistoni-~ulistorum, and uictoria-lebanonensis are sibling species

of Drosophila.

D N A HYBRIDIZATION IN DROSOPHILA

1 59

Days Cf incubation

FIGURE 2.-The rate of reassociation of the slow annealing DNA fraction. One mg of 3H- melunoguster DNA (688 cpm/pg; slow fraction, prepared as outlined in MATERIALS AND METHODS)

was placed in 1 ml of 3>< SSC with 5-7 filters containing from 2.8-3.7 mg of homologous, non- fractionated filterbound DNA, and was incubated with mild shaking in silicone-treated vials at 65°C for the specified number of days.

ments from the comparison species. It was predicted that E. coli DNA would not compete with the annealing of Drosophila DNA. Figure 3 shows the competition interactions of several DNAs with that of D. melanogaster. Table 2 presents the DNA comparisons of the species selected to represent the sections and subgenera of Drosophila. Six species were used as reference species, that is, they were used as the source of radioactive DNA in different competition experiments. These six species were selected to represent the respective species groups. This generated a large number of species comparisons and provided an opportunity to check the internal consistency of the results (see DISCUSSION). Table 3 presents DNA com- parisons of a group of closely related flies, six species in the melanica subgroup.

b. Similarity of the slow annealing DNA fraction. The question naturally arises as to how the fast annealing DNA relationships compare with those based on the slow annealing DNA fraction. The nucleotide similarity relationships using the slow annealing DNA of five species were obtained and compared with the fast annealing DNA relationships. The species relationships based on the slow annealing fractions do not differ significantly from those based on the fast anneal- ing fractions (Table 4). Since large quantities of DNA were needed, the direct annealing method rather than the competition method was used. E . coli DNA was again used as a nonannealing control.

60 T. D. ENTINGH

151 II A

A

A A A

simulans 0

0

0

, h. melanogastar.

50 100 150 200

p g unlabeled competitor DNA added

FIGURE 3.--Competition of DNA from different species in the reaction of 3H-melanogaster DNA fragments and filterbound melanogaster DNA. Increasing amounts of unlabeled DNA from various species were incubated with 5 pg 3H-melanogaster DNA (1576 cpm/pg) and 50 pg filterbound homologous DNA in 1.15 ml of 3>< SSC at 65°C for 16 hours.

TABLE 3

The relatedness of the DNAs of several species in the Melanica species group of the genus Drosophila, as measured by competition interactions

Source of commtitor DNA

D. pengi D. micromelanica D. nigromelanica D. euronotus D. paramelanica D. mlanica

Source of radioactive fragments D. micromelanica D. paramelanica (1,112 c p m / d (1,816 cpm/Pg)

67.7 66.4 100.0 72.8 77.2 71.3 73.3 88.2 71.2 100.0 70.2 92.1

Conditions of incubation as in legend to Figure 3. The percentage of radioactive fragments bound in the competition interactions with homologous DNA is used as a reference of 100% similarity. Values are averages of four samples with a mean standard error of 3.0% (range of standard error 2.M.9%).

D N A HYBRIDIZATION IN DROSOPHILA

TABLE 4

Comparison of species relationships based on hybridization studies with fast- and slow-annealing DNAs

61

Species compared with D . melanogarter

Nucleotide sequence Nucleotide sequence similarity with the slmilarity with the

Slow DNA fraction' Fast DNA fraction+

D. simulam D. uictoria D. uirilis D. funebris

78.2, 80.1 59.3, 56.9 42.2, 37.8 27.5, 20.2

81.5 69.3 42.6 34.2

* Measured by the direct annealing method. One mg 3H-melanogaster D N A fragments (slow fraction, prepared as in METHODS AND MATERIALS; 688 cpm/pg) was added to 1 ml3x SSC with 5-7 filters containing 2.4-3.9 mg of nonfractionated DNA from various species. This mixture was incubated in silicone-treated vials for 20 days at 65°C. + Measured by the direct competition method (see Figure 3). Values are taken from Table 4.

3. Hybridization studies with DNAs containing varying amounts of hetero- chromatin:

The exact nature and function of the fast annealing DNA is unknown. If the fast annealing DNA sequences originated solely from the heterochromatic por- tions of the genome, the amount of fast annealing DNA should be proportional to the amount of heterochromatin present. To test this hypothesis, DNA was extracted from D. melanogaster stocks containing varying amounts of X and Y heterochromatin, and their amount of fast annealing DNA was determined

Chromosomes

-- - I

I---

Description

.XlO males

Z DNA

21.06 2

2.15

20.6 +

2:65

21.5 +

FIGURE 4.-The nature of D. melanogaster stocks (with varying amounts of heterochromatin) used and their repetitive DNA content. Fifty pg of filterbound D N A from the above stocks \vas placed in 1.15 ml of 3>< SSC containing 5 pg of homologous 3H-DNA fragments (730-1160 cpm/pg) and incubated at 65°C for 16 hours. The values are means of five samples +- one stand- ard error.

62 T . D. E N T I N G H

(Figure 4 ) . The results do not support the idea that the fast annealing sequences originate solely from all the heterochromatic portions of the Drosophila genome.

DISCUSSION

Certain properties of DNA reassociation must be considered before the relation- ships found among the Drosophila species can be evaluated. First, the degree of resolution of the technique can be varied with the experimental conditions used (WALKER and MCCLAREN 1965; MARTIN and HOYER 1966). Thus the present data do not give absolute measures of nucleotide sequence similarity. But as long as incubation conditions are held constant, the technique gives reliable indications of relative nucleotide sequence similarity.

Second, in all higher organisms studied to date, a class of rapidly annealing DNA has been found in addition to slow annealing DNA sequences. The evidence that the rapidly annealing DNA consists of frequently repeated nucleotide se- quences has been reviewed in detail by BRITTEN and KOHNE (1968). The func- tion of this fast annealing DNA is unknown and the manner in which it in- fluences the processes of evolution and speciation can only be speculated upon. Although the present results (Figure 4) indicate that the amount of nucleolar organizer heterochromatin present bears no direct relationship to the amount of rapidly annealing DNA, the possibility remains that the fast annealing DNA is related to a specific chromosome structure.

The primary purpose of this study was to evaluate the type of information ob- tained by comparing a large number of genes through the DNA hybridization technique. Before discussing the major results, inferences made from some pre- vious hybridization studies should be pointed out. From vertebrate species, it ap- pears that similarities in polynucleotide sequences are related to the time at which the lines of the vertebrate organisms examined diverged from one another (HOY- ER et al. 1965). Similarly, SALTHE and KAPLAN (1966) using the microcomple- ment fixation technique, conclude that changes in the immunological properties of muscle-type lactate dehydrogenase in different amphibians show a linear re- lationship with time of phylogenetic divergence, as does the number of amino acid substitutions of the hemoglobin chains of mammals (ZUCKERKANDL and PAULING 1965) and of the cytochrome proteins of different organisms (MAR- GOLIASH and SMITH 1965). The term “phyletic distance” is often used to describe the species relationships inferred from the above types of data. As will be evident from the discussion below. one must be very cautious in assuming that “phyletic distance” is anything more than an inference from the data.

From the Drosophila data in Table 2 a number of inconsistencies become ap- parent. For example, there is a 29% difference in fast annealing nucleotide se- quence similarity between D. uictoria and D. virilis when D. melanogaster is used as the reference species, but only an 1 1 % difference when willistoni is used as the reference species. Numerous analogous discrepancies are cited in Table 5. The phylogenetic position of the reference species “viewing” the remaining species is very important in determining the relationships observed among the remaining

DNA HYBRIDIZATION IN DROSOPHILA 63

TABLE 5

Percentage differences between pairs of species depending upon the reference species used"

Reference species D. Species pair melanogaster uirilis hydeii willistoni

D. simulans

D. borealis 42.5 . . . 25.8 . . .

D. piulistorum

D. uicioria

D. paulisiorum

D. fulvnnacula

D. victoria

D. borealis

D. uictoria

D. fuluamacula

D. borealis

D. fulvamacula

23.7 8.0

52.0 58.9

30.3 13.4

41.3

11.0

35.2

19.1

* The percentages are calculated from the data given in Table 2.

species. The discrepancies which arise when a simple relationship between nucleotide sequence similarity and time of divergence is assumed are not sur- prising. It is unlikely that the evolutionary divergence found between various species is due to changes in the same DNA segment in the same way and to the same extent in the various lineages. As a result different relationships would appear among the species, depending on which species was used as the reference species. DNA comparisons do indicate the degree of similarity, but they are im- precise indicators of phylogeny, i.e., they are phenetic rather than phylogenetic comparisons. The observation that DNA comparisons (and comparisons of ho- mologous proteins) are phenetic comparisons has been made previously by THROCKMORTON ( 1968b).

The above statement can be tested by comparing the data on DNA comparisons with other types of data obtained from the same Drosophila species. THROCK- MORTON (1968a) has coded phenetically 60 anatomical characters from a number of different Drosophila species (a number of different states can exist for each character). The character states are coded descriptively and weighted equally. No hypothesis as to the evolutionary sequence of the character states is implied. The percentage of morhopological character states that two species shared was calculated and compared with the percentage of repetitive nucleotide sequences that these same species shared. Rigure 5 shows the regression of the percentage of character atates shared by two species on the percentage of repetitive nucleo- tide sequences shared. The high correlation coefficient (r = 0.79) between mcleo-

64 T. D. ENTINGH

3

% nucleotide sequences shared

FIGURE ti.-Regression of the percentage of morphological character states shared by two species (from THROCHMORTON) on the relative percentage of nucleotides these two species share (from Table 2).

tide sequences and morphological character states supports the view that DNA comparisons are phenetic comparisons. Thus for the genus Drosophila generally, it appears that DNA comparisons between species yield overall similarity values analogous to similarity values which can be obtained by comparing a large num- ber of equally weighted morphological character states of species.

From Figure 5 one infers that the greater the morphological similarity between two species the greater the repetitive nucleotide sequence similarity. Thus one would predict that a greater degree of nucleotide sequence similarity would exist between sibling species than between morphologically distinct species. The nu- cleotide sequence similarities between the sib1in.g species used in this study are summarized in Table 6. It can be seen that a high percentage (an average of 89% ) of nucleotide sequence similarity exists between sibling species. HUBBY and THROCKMORTON (1 968) using electrophoretic techniques studied an average of 18 proteins per species of nine triads of species in the genus Drosophila. Each

D N A HYBRIDIZATION IN DROSOPHILA

TABLE 6

Comparisons of the genetic similarity between sibling species on the basis of protein analysis and DNA hybridization studies

65

Sibling species

Percentage of Percentage of proteins in common repetitive ,nucleotide

sequences m common (maximum estimate) *

D. melanica

D. paramelanica

D. melanogaster

D. simulans

D. willistoni

D. paulistorum

D. victoria

D. lebanonensis

31.6

52.9

22.5

85.7

92.1

81.5

88.7

92.7

~ ~ ~~~~

* HUBBY and THROCKMORTON (1968).

triad consisted of a pair of sibling species and a related but morphologically dis- tinct species. Their results show that, on the average, sibling species shared pro- teins with the same electrophoretic mobilities 18% of the time. Their results also indicate that a whole range of genetic similarity (from 85.7 to 22.5%) may exist between sibling species. Table 6 compares the percentage similarity between four sibling species pairs based on the protein data of HUBBY and THROCKMORTON (1968) and the nucleotide sequence data. More comparisons are necessary, but it appears that an analysis of the electrophoretic mobility of proteins is a more sensitive technique for determining genetic divergence than an analysis of DNA sequence similarity using the hybridization technique. DNA hybridization studies cannot detect single base changes, whereas these might be detected in an analysis of proteins (providing the base change caused an electrophoretic change). In the present experiments the hybridization technique could not distinguish between two strains of D. uirilis (1801.1 and 1736.7). Melting curves of the reassociated duplexes of these strains might possibly distinguish the two strains. One would predict that an analysis of proteins would reveal divergence, since genetic poly- morphism has been found in different populations of Drosophila species (HUBBY and LEWONTIN 1966; LEWONTIN and HUBBY 1966).

It can also be noted from Table 2 that the reciprocal comparisons which were made (melanogaster and virilis; hydeii and willistoni; paramelanica and me- lanica) all agreed. This would be the case only if these species have the same total complement of DNA. Whether this is true for all the Drosophila species needs to be determined from more reciprocal comparisons. The fact that all of the species tested have approximately 20% repetitive DNA is very interesting, although the implications of this are not clear.

All of the above comparisons were made with the fast annealing DNA frac-

GG T. D. ENTINGH

tions. Although the present method of separating fast and slow annealing DNA is very crude when compared to the hydroxyapatite fractionation procedures of BRITTEN and KOHNE (1968), the fast annealing fraction is apparently a repre- sentative sample of the genome, since the species relationships based on the fast annealing fractions do not differ significantly from those based on the slow an- nealing fractions (Table 4). In contrast, different species of mammals are more divergent when the slow annealing DNA is compared instead of the fast anneal- ing DNA. For example, there appear to be few, if any, common slow annealing sequences between the mouse and rat; although these organisms share over 50% of their fast annealing sequences (BRITTEN and KOHNE 1966). One interpreta- tion of these data is that the slow annealing sequences change at a much faster rate than the fast annealing sequences. Another interpretation is inherent in the DNA hybridization technique. In order for two unique sequences to reassociate, base complementarity must be very precise; but when a number of similar se- quences are present, a certain amount of base mispairing may exist without pre- cluding the reassociation of these sequences. Thermal stability studies of the re- associated duplexes should be helpful in distinguishing between the above possi- bilities, and they might provide information as to why the slow and fast fractions reveal the same species relationships in the case of Drosophila.

One interesting feature of these results is the magnitude of the differences found between the individual species. A bacterial genus such as Bacillus (DUBNAU et al. 1965) can contain species even more diverse than the Drosophila species, but the taxon of higher plants showing this degree of diversity seems to be the family (BOLTON et al. 1965). In mammals, the taxon showing this degree of di- versity seems to be the Order (BRITTEN and KOHNE 1966). This difference in magnitude of divergence between higher plants and animals and the genus Dro- sophila is probably due to differences in rates of divergence, as seen by the ap- pearance of new genera and orders and by an increase in the number of species. Hopefully, a correlation of various patterns of speciation and degrees of nucleo- tide substitutions will eventually permit a determination of the rates of diver- gence within various groups of organisms.

I would like to express my gratitude to Dr. J. L. HUBBY €or many helpful discussions, and to Dr. L. 13. THROCKMORTON for criticizing the manuscript and for his contribution to Figure 5 .

SUMMARY

Some experimental parameters of intraspecific and interspecific DNA/DNA duplex formation in the genus Drosophila were investigated. The genomes of eighteen Drosophila species were found to consist of approximately 20% fast annealing DNA, in addition to the slow annealing DNA. Investigations into the possible nature of D. mehogaster fast annealing DNA revealed that it does not originate solely from the heterochromatic portions of the genome. The relative percentage of fast annealing DNA sequences shared among these eighteen species and the percentage of slow annealing sequences shared among five of them were determined. The species relationships based on the respective DNA fractions did

D N A HYBRIDIZATION IN DROSOPHILA 67

not differ significantly. A simple relationship between the percentage of nucleo- tide sequences shared between two species and the time of their divergence was not found, and the importance of the phylogenetic position of the “reference” species in determining the relationships that will be observed among the remain- ing species was illustrated. It was found that DNA comparisons between species yield overall similarity values which are analogous to similarity values that can be obtained by comparing a large number of equally weighted morphological characters. Thus DNA comparisons are phenetic comparisons. The applications of the DNA hybridization technique and the relevance of DNA hybridization results in providing a better understanding of species relationships and evolu- tionary processes were discussed.

LITERATURE CITED

BOLTON, E. T., R. J. BRITTEN, G. B. COWIE, R. B. ROBERTS, P. SZAFRANSRI and M. J. WARING, Carnegie Inst. Washington Annual Report 1964-1965, pp. 313-333.

BRITTEN, R. J. and D. E. KOHNE, 1966 Carnegie Inst. Washington Annual Report 1965-1966, pp 78-106. -, 1968 Repeated sequencesin DNA. Science 161: 529-540.

CARPENTER, J. M., 1950 A new semi-synthetic food medium for Drosophila, Drosophila Inform. Serv. 24: 96-97.

DENHARDT, D. T., 1966 A membrane-filter technique for the detection of complementary DNA. Biochem. Biophys. Res. Commun. 23: 641-646.

DISCHE, Z., 1944 Two characteristic and sensitive color reactions between sulfhydryl com- pounds and thymonucleic acid. Proc. Soc. Exptl. Biol. Med. 55: 217-218.

DUBNAU, D., I. SMITH, P. MORRELL and J. MARMUR, 1965 Gene conservation in Bacillus species. I. Conserved genetic and base sequence homologies. Proc. Natl. Acad. Sci. US. 54: 491498.

GILLESPIE, D. and S. SPIEGELMAN, 1965 A quantitative assay for DNA-RNA hybrids with DNA immobilized on a membrane. J. Mol. Biol. 12 : 829-842.

HOYER, B. H., E. T. BOLTON, B. MCCARTHY and R. ROBERTS, 1965 The evolution of polynucleo- tides. pp. 481-590. In: Evolving Genes and Proteins. Edited by V. BRYSON and H. J. VOGEL. Academic Press, New York.

Studies of nucleic acid interactions using DNA-agar. pp. 425-478. In: Molecular Genetics, vol. 11. Edited by J. H. TAYLOR. Academic Press, New York.

HUBBY, J. L. and R. C. LEWONTIN, 1966 A molecular approach to the study of genic heterozy- gosity in natural populations. I. The number of alleles at different loci in Drosophila pseudo- obscura. Genetics 54: 577-594.

Protein differences in Drosophila. 11. Compara- tive species genetics and evolutionary problems. Genetics 52: 203-215. -, 1968 Pro- tein differences in Drosophila. IV. A study of sibling species. Am. Naturalist 102 : 193-205.

Magnitude of interspecific nucleotide sequence variabil- ity in Drosophila. Genetics 60: 303-322.

A molecular approach to the study of genic heterozy- gosity in natural populations. 11. Amount of variation and degree of heterozygosity in natural populations of Drosophila pseudoobscura. Genetics 54: 595-609.

Structural and functional aspects of cytochrome c in rela- tion to evolution. pp. 221-242. In: Evolving Genes and Proteins. Edited by V. BRYSON and H. J. VOGEL. Academic Press, New York.

1965

HOYER, B. H. and R. B. ROBERTS, 1968

HUBBY, J. L. and L. H. THROCKMORTON, 1965

LAIRD, C. D. and B. J. MCCARTHY, 1968

LEWONTIN, R. C. and J. L. HUBBY, 1966

MARGOLIASH, E. and E. SMITH, 1965

68 T. D. ENTINGH

MARMUR, J., 1961

MARTIN, M. A. and B. H. HOYER, 1966

MAYR, E., 1963 MCCARTHY, B. J. and E. T. BOLTON, 1963

PATTERSON, J. T. and W. S. STONE, 1952

RITOSSA, F. M. and S. SPIEGELMAN, 1965

A procedure for the isolation of deoxyribonucleic acid from microorganisms.

Thermal stabilities and species specificities of reannealed J. Mol. Biol. 3: 208-218.

animal DNAs. Biochemistry 5 : 2706-2713.

Animal Species and Evolution. Belknap, Cambridge, Mass An approach to the measurement of genetic related-

ness among organisms. Proc. Natl. Acad. Sci. U.S. 50: 1561M. Evolution in the Genus Drosophila. Macmillan, New

York. Localization of DNA complementary to ribosomal

RNA in the nucleolus organizer region of Drosophila melanogaster. Proc. Natl. Acad. Sci. U.S. 53: 737-745.

Immunology and rates of enzyme evolution in the amphibia in relation to the origins of certain taxa. Evolution 20: 603-616.

The homologies of the chromosome elements in the genus Drosophila. Genetics 26: 51 7-541.

The problem of phylogeny in the genus Drosophila. Studies in Genetics 11. Univ. Texas Publ. 6205: 207-343. -, 1962b The use of biochemical char- acteristics for the study of problems of taxonomy and evolution in the genus Drosophila. Studies in Genetics 11. Univ. Texas Publ. 6205: 415-587. -, 1968a Concordance and discordance of taxonomic characteristics in Drosophila classification. System. Zml. 4 : 307- 341. -, 1968b Biochemistry and taxonomy. Ann. Rev. Entomol. 13: 99-114.

Specific duplex formation in uitro of mammalian DNA. J. Mol. Biol. 12: 394-409.

Evolutionary divergence and convergence in proteins. pp. 97-166. In: Euoluing Genes and Proteins. Edited by V. BRYSON and H. J. VOGEL. Aca- demic Press, New York.

SALTHE, S. N. and N. 0. KAPLAN, 1966

STURTEVANT, A. H. and E. NOVITSKI, 1941

THROCKMORTON, L. H., 1962a

WALKER, P. B. and A. MCCLAREN, 1965

ZUCKERKANDL, E. and L. PAULING, 1965